What Didn’t Work: Unsuccessful UDRP Complaints in 2016

This article analyzes UDRP decisions in favor of the respondent domain owner in 2016. What factors sunk complainants and what fact patterns favored respondents?

Data set

The data was manually coded from the National Arbitration Forum (“NAF”) searchable case database, and taken from the aggregate reporting provided by the World Intellectual Property Organization (“WIPO”).

Both organizations have strikingly similar outcomes: roughly 90% of complaints succeeded in UDRP claims. Of those, a bit more than two percent of the WIPO decisions took the baffling step of canceling the domain name instead of seeking its transfer, putting the domains back in the general registration pool, where the complainant has no better chance than anyone else of getting the domain name. The one strategic reason to file for cancellation may be where the domain name included marks from both the complainant and a third party, where the complainant may not be able to secure transfer.

Outcome summaries are presented below in a table and as a pie chart.

Outcome Cases Percentage Cases Percentage
Cancellation 3 0.20% 51 2.16%
Transfer 1300 87.72% 2133 90.23%
Denied 179 12.08% 179 7.57%
Total 1482 100% 2363 100%


NAF.png WIPO.png

Since the overall outcomes are so similar and time is not unlimited, we focused only on NAF results for a more detailed breakdown.


We analyzed each victory for a respondent based on the grounds cited by the UDRP panel. Many decisions had more than one ground cited, so the total count of issues exceeded the raw number of decisions. There were some clear patterns in the results, reviewed below.


The Complainant Lost a Prior UDRP or Court Proceeding

Complainants always lost where there was res judicata applied. It would seem self-evident that a prior loss in a UDRP proceeding or domain-related litigation would preclude winning a UDRP in the future, since the UDRP is primarily focused on the rights of the complainant and conduct of the respondent at the time of registration. Several complainants tried anyway and predictably failed.

The Complainant Does Not Have Priority

This was the second clear loser for complainants and was much more common than res judicata. Even though the UDRP requires evidence of bad faith registration and use to prevail, about 25% of the losing complainants filed complaints where the domain owner acquired the domain name in question before the complainant had trademark rights. UDRP decisions uniformly hold that registration in these circumstances cannot be in bad faith (even if use can) and also generally hold that the respondent has rights or legitimate interests in the domain name.

The typical fact pattern involved the complainant trying to capture the .com domain name corresponding to the mark where the domain name was registered before the complainant adopted or first sought to register its mark. The complainants quite reasonably want to get the domain name, often attempting (unsuccessfully) to purchase it from the owner, and then quite unreasonably used a UDRP proceeding to attempt to get their hands on it. Of course, the registration could not have been in “bad faith” where the complainant had no rights at the time of registration! This fact pattern is most likely to lead not only to a quick and firm rejection of the complainant’s claims, but also to a “reverse domain name hijacking” finding by the UDRP panel.

Where the domain name was registered prior to the complainant’s first trademark rights, but has been used in bad faith since (e.g. for pay-per-click ads for competitive goods or the like), the UDRP is just not the right tool. Complainants would have a perfectly good ACPA claim on the same facts. Whether these complainants were unaware of the differences in the claim or simply wanted to avoid the cost of federal court litigation is not clear, but filing a lawsuit isn’t wildly more expensive than filing a UDRP, and it’s easy to leverage a clearly winning ACPA claim into a quick settlement, where filing a clear-loser UDRP complaint is little better than burning money.

The Complainant Did Not Adequately Pleading (or Did Not Have) Common-Law Rights

Poorly-supported claims of common-law rights were another common loser, totaling roughly 20% of the UDRP losses by complainants. The typical fact pattern was a small or new business, or an individual claiming trademark rights in their name – often filing the complaint without a representative – filing a complaint that provided little more than a business name registration or a Facebook page as evidence of common-law rights. Many of these complaints, like those in the “priority” section above, sought to take high-value domain names, often those consisting solely of a combination of generic terms. To be clear, common-law claims in general were not problematic for complainants: common-law complaints backed by significant evidence of sales, advertising expenditures, and public recognition like press coverage had no more issues than registration-based complaints in providing rights under 4(a)(i).

The Complaint Targeted a Criticism Site

Complainants have largely figured out that the UDRP is not a great tool for trying to shut down critical sites that are not actively trying to confuse visitors. While complainants occasionally prevail – there is a bit of a split in the UDRP case law on the point – 5% of losses still related to criticism sites.

The Complaint Used the UDRP to Address a Business Dispute

About 15% of disputes failed because they fell outside the scope of the UDRP. Most were situations where the parties had some relationship, e.g. a former employee or officer of the complainant registered the domain name in their name while working for the complainant, and the complainant tried to use a UDRP claim to get it back. Panels tend to reject these claims as fundamentally contract or agency questions rather than questions within the scope of the UDRP.

The Complaint Addressed a Reseller

About 5% of the failed claims related to resellers, where the panel found that, between the nature of the domain name and the site, the registrant was attempting to operate a legitimate and non-confusing resale operation.

Other issues

About 10% of complaints also foundered on the “rights or legitimate interests” inquiry (largely cases where the complainant ran a site that was not competitive with the trademark owner’s business) or “bad faith” inquiry (generally where an offer to sell in response to a purchase inquiry from the domain owner was used as the sole proof of bad faith, and wasn’t enough, by itself, to show the registrant’s initial bad intent). Five of these involved successful claims that the registrant simply registered generic English terms that happened to match the complainant’s trademark, which was a generic English term for some goods but was used for unrelated goods. These defenses were most successful when the respondent had a huge portfolio of such domains, and didn’t appear to be targeting the complainant’s business (or the marks of other brand owners) in other ways. Interestingly, the respondent prevailed in a couple of proceedings even though an automated pay-per-click ad service put up ads competitive with the complainant; the panels refused to impute bad faith registration even though there was evidence of bad faith use. Those decisions were a bit of a departure from the norm.

A handful of proceedings were lost on other formalities grounds, most often where a complaint related to a domain name that included the marks of two companies, only one of whom was a party to the proceeding.


We hope this was a useful double-check for UDRP complainants prior to filing, and a useful guide to analyzing potential claim weaknesses for respondents in domain name dispute proceedings.

PYROSIL – Anatomy of an Office Action Response #4

Anatomy of an Office Action Response

Episode 4 – PYROSIL

What is this series?

We break down a recently-filed, successful Office Action Response, looking at the case law, evidence, and strategic decisions that made it a success.

Today’s post focuses on strategies in responding to a likelihood of confusion refusal featuring identical marks in different classes.

The Office Action – PYROSIL

The applicant’s filing for PYROSIL, Ser. No. 86912816, for a variety of products and services including silicon and other chemical precursors (Class 1), fuel and dust absorbing and binding compounds (Class 4), metered fuel pumps (Class 9), and chemical coating services (Class 44), and actually used for a silicon oxide coating and coating process, received a citation to a prior pending application for PYROSIL for paints and primers used in the aerospace industry (Class 2), Ser. No. 79179408. Both applications are now registered.

Because the prior PYROSIL application was still pending, the Office Action did not provide (and did not need to provide) arguments to support the refusal.

The correspondent, Otto Lee of Intellectual Property Law Group LLP, filed an Office Action Response arguing against the refusal. The prior applications’ limitation to the aerospace industry was a key: it limited the customer base of the cited mark, and the applicant supplemented the meaning of “industrial” products in its description with extrinsic information about the product and its customers. The Response emphasized that the products offered under both marks were sold to different and non-overlapping classes of sophisticated customers in distinct manufacturing industries, and were necessarily sold through different trade channels. While the applicant’s products are focused on adhesive promotion coatings, e.g. to increase print quality, the applicant’s website references automotive uses and neither the description of goods in the application nor the supplementary materials carve out or disavow aerospace or military use. That’s certainly a desirable outcome for the applicant, and a good strategic reason for counsel relying on extrinsic evidence rather than narrowing the applicant’s description of goods to exclude the fields.

Counsel also pointed to another out-of-class registration for PYROSIL, Reg. No. 2546215, for thermocouples in Class 9, noting that it was also sold to a niche group of professional buyers and could co-exist without confusion. This data point is valuable under the 6th du Pont factor, and arguably also restricts the prior registrant’s rights under the 11th du Pont factor as well.

The response made a variety of arguments about the lack relationship between the goods, ultimately successfully. The applicant could have pointed to registry evidence of co-existence of similar products, both in industrial silicon and fuels. In re Thor Tech, Inc., 113 USPQ2d 1546 (TTAB 2015). The following table, from TM TKO’s ThorCheck prosecution tool, identifies a number of instances where identical or highly similar marks co-exist without apparent confusion for exactly the sorts of goods.

Owner Goods/Services Mark Mark Goods/Services Owner
Regal Chemical Company 001 silicone emulsions for control and suppression of foam in aqueous solutions REGAL
Reg: 3842060
Serial: 77461042
Reg: 1706125
Serial: 74196056
002 interior paint Columbia Insurance Company
Dow Corning Corporation 001 polycrystalline silicon, polysilicon, silanes, and silicon source chemicals used in manufacturing silicone ingots and wafers for semiconductor devices (…) HSC
Reg: 3734230
Serial: 77761283
Reg: 3469539
Serial: 78818952
002 paints Superior Products International Ii, Inc.
Emerald Performance Materials, Llc 001 specialty and fine chemicals, resins and polymers, namely, (…) silicone-based additives in the nature of low molecular weight silicone-based oligomers, (…) all for use in the manufacture of (various products) EMERALD
Reg: 4344989
Serial: 85324867
Reg: 4686712
Serial: 85980802
002 paint thinners Emerald Services, Inc.
Petroleum Marketing Analysis, Inc., dba Aura Oil & Lamp Creations 004 lamp fuel AURA
Reg: 4007891
Serial: 77300540
Reg: 3220824
Serial: 78753154
002 exterior paint Columbia Insurance Company
Chevron Intellectual Property Llc 004 lubricating-oils REGAL
Reg: 0088032
Serial: 71058463
Reg: 1706125
Serial: 74196056
002 interior paint Columbia Insurance Company
Champion Brands, LLC 004 stove and lantern fuel HI-TECH
Reg: 2256277
Serial: 75239061
Reg: 1295391
Serial: 73404876
002 spray paint Seymour Of Sycamore Inc.
Tiara Brands, Inc. 004 gasoline CROWN
Reg: 2069133
Serial: 75081978
Reg: 1334990
Serial: 73493222
002 paint thinner Packaging Service Co., Inc.
West Penn Oil Company, Inc. 004 gasoline EMBLEM
Reg: 0330234
Serial: 71367514
Reg: 4732801
Serial: 86190094
002 paints The Valspar Corporation
Regal Beloit America, Inc. 004 lubricating oils and greases WAVERLY
Reg: 4489586
Serial: 85889529
Reg: 2340120
Serial: 75541752
002 house paint Icon De Holdin GS Llc
Sterno Products, Llc 004 chafing dish fuel HOT SPOT
Reg: 4500511
Serial: 85835539
Reg: 1311806
Serial: 73403836
002 spray paint Seymour Of Sycamore Inc.
Charles Carney

dba Pro Tech Lubricants

004 diesel oil PROTECH
Reg: 4661852
Serial: 86043852
Reg: 4847575
Serial: 86424118
002 architectural paints Swimc, Inc.
Micro Powders, Inc. 004 wax emulsions and wax powders for industrial applications AQUAMATTE
Reg: 4209568
Serial: 85540685
Reg: 2091731
Serial: 74642767
002 interior paint Swimc, Inc.

Following the successful Office Action Response, the application will be published for opposition in March 2017.

Immigration-Related Trademark Filings

Immigration has been in the news frequently over the past few years, and increasingly so in 2017. We took a quick look at whether the rate of immigration-related filings at the USPTO has changed markedly over the years in a way that reflected the political salience of the issue. The search was limited to a handful of immigration-related terms in the descriptions of goods and services, and is not anywhere close to a definitive count of all marks filed for immigration-related products or services. Just for one example, the search does not even try include broader descriptions of services, like “legal services,” that could encompass immigration-related services, like immigration law.

Recent years have certainly seen an uptick in immigration-related applications, surging from fifty a year in the 2000s to around 150 per year in the last three years. The following chart summarizes the increase in average yearly application rates.

immigration by year.png

While the trademark application rate for specifically immigration-related products and services has increased significantly, it has not increased in relation to the increase in overall trademark filings. The following chart normalizes both immigration-related filing activity and overall filing activity. While not exactly in lockstep, the ratios are quite similar; if anything, overall filing volumes appear to have jumped more than immigration-related filings.

immigration normalized.png

Charities’ Trademark Filings

Like any other venture, charities have trademark rights. Charities’ names and logos are important assets, and consumer confusion (especially donation-related scams) can be a significant problem for many charities. With that in mind, we took a look at how the twenty-five largest charities in the US treat their trademark portfolios.

There is as much variation in the charities’ trademark protection strategies as there is in the organizations’ charitable missions. Some, like the United Way, St. Jude, and Cru, are very active applicants with numerous trademark registrations and pending applications; others, like the Task Force for Global Health and the Patient Access Network Foundation, have little to no registry presence. There were no obvious correlations between trademark filing decisions and the nature of the charity, its geographic focus, or its religious affiliation or lack thereof.

The graph below is normalized: blue shows the charity’s revenue, red total US trademark filings, and yellow 2016 trademark filings.


For the curious, the non-normalized data follows: it includes not only the three categories charted above but also the “expense ratio” from CharityNavigator.org — the percentage of revenue spent on the stated charitable goal, at least for those organizations required to publicly report that data. CharityNavigator.org has many additional metrics that normalize expense ratios across different types of organizations, which may necessitate very different cost structures.

Organization Revenue Active 2016 Expense Ratio
United Way Worldwide $3,708.00 174 16 91.60%
Task Force for Global Health $3,154 0 0 not rated
Feeding America $2,150 49 3 98.50%
Salvation Army $1,904 77 2 not rated
YMCA of the USA $1,202 74 1 not rated
St. Jude Children’s Research Hospital $1,181 210 14 69.20%
Food for the Poor $1,156 7 0 95.70%
Boys & Girls Clubs of America $923 32 0 81.80%
Catholic Charities USA $921 2 0 78.20%
Goodwill Industries International $902 82 7 not rated
Habitat for Humanity International $829 40 1 81.90%
World Vision $825 40 2 83.90%
American Cancer Society $810 79 6 59.90%
Patient Access Network Foundation $801 1 0 not rated
Compassion International $799 22 2 83.10%
Direct Relief $775 5 0 99.20%
Americares Foundation $740 32 2 97.90%
Lutheran Services in America $723 7 1 not rated
Nature Conservancy $646 35 4 71.20%
American Heart Association $634 97 11 78.70%
American National Red Cross $624 46 0 90.10%
Samaritan’s Purse $565 19 0 88.30%
MAP International $545 15 0 99.20%
Step Up for Students $521 0 0 98.90%
Cru $514 217 6 not rated

Marijuana-related trademark filing trends

The US Patent & Trademark Office will not register trademarks for drugs that are regulated under the Controlled Substances Act or for related paraphernalia. 21 U.S.C. §§801-971; TMEP § 907. That does not stop applicants from applying for a wide range of marijuana-related products and services, and many products that are not the drug itself nor paraphernalia are registrable. Direct registration protection for marijuana products is available via state registries. This blog post looks at filing trends in the USPTO’s federal registry.

California legalized medical marijuana in 1996, with several Oregon (1998) and Main (1999) following suit. Six states legalized medical marijuana in the 2000s and nine more did so in the 2010s. Recreational legalization began with Colorado and Washington in 2012 and has since expanded to nine more states.


Marijuana-related trademark filings on the federal level did not exactly match up with state-level legalization trends, possibly because legalization tends to require slow and bureaucratic rulemaking. Federal trademark filings started to accelerate in 2009, and really skyrocketed in 2014.

Most federal trademark filings are focused on a handful of classes. The chart below shows filings where the mark, the mark description, or the description of goods contains certain “single-purpose” marijuana-related terms. For instance, “cannabis” was included as a search term but not “joint,” whose primary use is not marijuana-related.

The key classes were:

  • Class 5 for medicines; these will presumably be almost uniformly refused under the Controlled Substances Act;
  • Class 25 for clothing; these “merchandising” or “messaging” products are generally accepted by the USPTO;
  • Classes 30-34, presumably for edibles; these also face an uphill battle under the CSA;
  • Class 35 for retail services, which will face CSA issues, and online review/recommendation services, which should be OK;
  • Class 41 for entertainment services, generally OK; and
  • Class 44 for medical services, largely on the medical marijuana side of the industry, and which generally face CSA issues.


As state-level legalization efforts continue, it is possible that Congress will amend the Controlled Substances Act to permit limited intellectual property protections for marijuana-related products and services in states where they are legal. Until then, applicants in the marijuana industry will have to seek as much federal trademark protection as they can for ancillary products or for merchandised goods, and rely on a patchwork of common-law rights and state trademark registrations to protect the brand identifies of their core products.

Anatomy of an Office Action Response #5


What is this series?

We break down a recently-filed, successful Office Action Response, looking at the case law, evidence, and strategic decisions that made it a success.

Today’s post focuses on an application that received a disclaimer request.


We are looking at a successful Office Action Response filed in the course of prosecuting an application to register the mark ORANGE TREE SAMPLES for autio and musical sound recordings and related software in International Class 32, Ser. No. 86958626. The applicant, Greg Schlaepfer, is represented by Philip Zender of Squire Patton Boggs in San Francisco.

In an Office Action issued on July 19, 2016, the Office found no confusingly similar marks but requested a disclaimer of the term SAMPLES, pointing both to external evidence and the use of “sample libraries” in the description of goods.

Reviewing the Office Action

While not noted by the Examining Attorney, there is considerable case law supporting the Office’s position. Use of a term in a description of goods or services is just one instance of the applicant’s own descriptive or generic use of the term in question, which is generally highly probative of the question of the strength of that term. See, e.g., In re Hunter Fan Co., 78 USPQ2d 1474, 1476 (TTAB 2006) (“applicant’s own use of the term ERGONOMIC … highlights the descriptive nature of this term …”).

Let’s look at the Office Action using TM TKO’s tools. Is “sample” a common term on the registry, indicating that it is not likely to serve as a source indicator?


TM TKO’s look at registry data also suggests that the term is fairly weak. Mousing over the graph for more detail indicates that “sample” is common in descriptions of goods in-class and, where it occurs in marks, it is often disclaimed or present in marks in Supplemental Register registrations.


Combined with the use of the term in the description of goods and the case law on the point, the disclaimer request in the Office Action looks well-supported.

The Office Action Response

The applicant saw no reason to push back, and agreed to the disclaimer in an Office Action Response filed on October 17, 2016. The application was published for opposition on December 13, 2016.

Design search and watch – new to TM TKO!

Design Search Guide

TM TKO now provides high-quality design search reports in addition to word mark searches. Go to the “Knockouts” page and pick the “Design Mark” tab. You will see our design search interface:


Entering a search

You need to enter at least one design element to search in either the “Design Description” or the “Design Codes” fields; these fields can be combined to further prioritize results.

The “Design Description” box is free-text, and searches what on TESS is the [de] (mark description) field. Each term is searched separately as a keyword, and results hitting on all keywords are prioritized over those matching fewer keywords.

The “Design Codes” tab auto-fills as you type, based on what in TESS would be the [dd] (design description) and [dc] (design code) fields. You can enter the six-digit design codes with or without periods, i.e. 050106 or 05.01.06, or you can also start typing “stump” to select the same design description and design code without having to reference the design search manual. Click on the entry you want or scroll down to the item and hit enter.


After selecting the design phrase and code, it will appear below the design search bar. Click “x” to remove any phrase.


As with design descriptions, you can enter more than one design code in a single search.

Enter Goods and Services

You need to enter at least one good or service to run a search.

Under “Goods and Services”, start typing a product or service. TM TKO auto-suggests the most common goods and services on the USPTO registry for you; we are able to provide the best results when you pick more common terms that accurately describe your client’s product or service. When you see one you like, click on it to add it to your list. There is no need to worry about class – TM TKO will automatically assign the class for each good or service you enter based on registry data, and you can add as few or as many terms as you like per search.

Here is the goods and services auto-suggester in action:


Here is what it looks like after you have selected goods and services:


This search would focus on International Classes 32 and 33, the classes for the two goods entered, and, as space allows, also highlight very similar marks in coordinated classes. As with design codes, click “x” to drop a given good or service from the search.

Presenting Results

Design search reports are a bit more traditional than our word mark search reports, focusing on a prioritized table of results. In-class results hitting on all design search criteria are prioritized before searches hitting on fewer criteria. If you searched using design codes, as you scroll down the results list, you will also see in-class results with similar but not identical design codes.

Unlike word mark search report, all design search results include the full mark description, design phrases, and design codes in the “Mark” field.


Tips and Tricks

– Limiting your search to a single design keyword will provide fewer results in-class and, for uncommon terms, more results in coordinated classes and for related goods.

For example, searching the Design Description field for “mastodon” for “luggage” yields only two results, both outside of International Class 18 (coordinated classes or statistically related goods). Doing the same search but with the “elephants, mammoths” [030301] design phrase and code added to the search yields nearly forty in-class results for those very similar animals, plus some additional results with other large savannah animals like giraffes and rhinos that have similar design codes. Including “mammoth” in the “Design Description” field further highlights results containing “mammoth” in the mark description, bringing them to the top of the chart above the “elephant” marks for easier review.

– Visual similarity between designs is often a judgment call. For most design search projects, we suggest running a separate search for each class of products or services that your client plans to offer, and to use both the most relevant design code(s) plus relevant keywords in the design description field to highlight the most relevant marks in the results table.

– To include abandoned applications and expired or canceled registrations, click the “Include all inactive filings” radio button near the “Knockout Search” button.

Any Questions?

Please don’t hesitate to email support@tmtko.com with any questions! We are happy to help.